JP2010185812A - Human body detecting device and urinal equipped with the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a human body detecting device capable of exactly recognizing the state of human behavior with a small amount of calculation. <P>SOLUTION: The human body detecting device utilizing a Doppler signal of a propagation wave reflected by an object to be sensed includes: a propagation wave transmission part for emitting the propagation wave toward the object; a propagation wave receiving part for receiving the propagation wave reflected by the object; and a Doppler signal generation part for generating the Doppler signal based on the propagation wave emitted by the propagation wave transmission part and the propagation wave received by the propagation wave receiving part; and has a behavior state determination part for analyzing a Doppler signal generated by this Doppler signal generation part by means of a particle filter and for determining the state of the human behavior based on a calculated distance to the object and moving speed. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a human body detection device, and more particularly, to a human body detection device using a Doppler signal of a propagation wave reflected by a detection object and a urinal provided with the human body detection device.

  Japanese Unexamined Patent Publication No. 9-80150 (Patent Document 1) describes a human body detection device. In this human body detection device, a microwave is emitted to the front of the toilet, the microwave reflected by the object is received, the power spectrum of the Doppler frequency signal is obtained, and the human action is based on the peak value of the power spectrum. The state is being judged.

  Japanese Patent Laying-Open No. 2006-214156 (Patent Document 2) describes a urinal washing device. In this urinal washing apparatus, a frequency filter that allows only a specific frequency band signal to pass is applied to the output signal of the microwave Doppler sensor, and the human behavior state is determined based on the output signal of the frequency filter.

  These human body detection devices using microwaves described in Japanese Patent Laid-Open No. 9-80150 (Patent Document 1) and Japanese Patent Application Laid-Open No. 2006-214156 (Patent Document 2) use infrared rays that are currently widely used. Unlike the human body detection device used, when applied to a urinal, etc., there is no need to provide a window for transmitting infrared rays in the urinal body, etc., so there are few restrictions on the conditions for installing the device, and the design is small. A toilet can be realized.

Japanese Patent Laid-Open No. 9-80150 JP 2006-214156 A

  However, in the human body detection device described in Japanese Patent Application Laid-Open No. 9-80150, since the power spectrum is obtained by performing Fourier transform on the detected Doppler signal, it is impossible to grasp the human action state in time series. There is a problem. That is, in the human body detection apparatus described in Japanese Patent Application Laid-Open No. 9-80150, the power spectrum is obtained by performing fast Fourier transform on 128 pieces of data obtained by sampling the Doppler signal over 0.64 sec at 5 msec intervals. ing. The peak frequency and amplitude information obtained from the power spectrum obtained in this way represents the average value over 0.64 seconds of sampled data, and immediately detects changes in human behavior to be detected. There is a problem that you can not.

  In addition, in the method of obtaining the power spectrum by fast Fourier transform, there is a problem that if the number of data to be sampled is reduced, the resolution of the obtained peak frequency is lowered. In addition, if the sampling interval is shortened and a large amount of data is acquired in a short time and fast Fourier transform is performed, the amount of calculation increases, and the power consumed by the human body detection device increases due to a large amount of calculations. is there.

  On the other hand, the frequency filter used in the device described in JP-A-2006-214156 can be realized by applying a digital filter to the detected Doppler signal. Sampling data is required. Further, there is a problem that it is not possible to sufficiently grasp the behavior of a person to be detected only by obtaining single data in the frequency domain using a single digital filter. That is, the information obtained from a single digital filter cannot identify whether the person to be detected has started walking from a stationary state or whether the person who has been walking is stationary.

  In order to solve this problem, it is conceivable to use a plurality of digital filters by narrowing the frequency range through which the digital filter passes, but in this case, as the number of digital filters increases, the computation of the digital filters A problem arises in that the amount increases and the calculation process cannot keep up.

  Accordingly, an object of the present invention is to provide a human body detection device capable of accurately grasping a human behavior state with a small amount of calculation and a urinal provided with the same.

  Another object of the present invention is to provide a human body detection device that can immediately detect a change in the behavior of a person to be detected, and a urinal equipped with the human body detection device.

  In order to solve the above-described problem, the invention according to claim 1 is a propagation wave transmitting unit that radiates a propagation wave toward a detection target, and a propagation wave that receives the propagation wave reflected by the detection target. In the human body detection device comprising: a reception unit; and a Doppler signal generation unit that generates a Doppler signal based on the propagation wave radiated by the propagation wave transmission unit and the propagation wave received by the propagation wave reception unit, Analyzed by a Doppler analysis physical model that combines a term that vibrates at a Doppler frequency that changes according to the moving speed of the object to be detected, a vibration amplitude, a DC component of the sensor signal, and a term in which the phase difference between the transmitted signal and reflected signal changes temporally A plurality of distance candidate values between the detection object and the human body detection device and a plurality of movement speed candidate values of the detection object are set while having a particle filter The estimation means for estimating a plurality of Doppler estimation signals using the distance candidate value and the moving speed candidate value, the amplitude values of the plurality of Doppler estimation signals and the Doppler signal are compared, and the difference is the smallest Calculated by the Doppler signal analysis unit that selects the distance candidate value and the moving speed candidate value corresponding to one Doppler estimation signal that causes an amplitude value error as the distance estimation value and the movement speed estimation value, and the Doppler signal analysis unit An action state determination unit that determines a person's action state based on a distance and a moving speed of the detection target.

  According to the present invention configured as described above, by using a Doppler analysis physical model in which a Doppler effect term and a beat term are combined using a particle filter, not only a moving speed of a person but also a detection object By estimating the distance between the human body detection device and the human body detection device, it is possible to determine a more accurate human behavioral state.

  According to a second aspect of the present invention, in the human body detection device according to the first aspect, the Doppler analysis physical model includes a Doppler effect term and a signal coefficient that multiplies a beat term, and the Doppler signal analysis unit includes: A plurality of amplitude coefficient candidate values indicating the amplification degree of the Doppler signal, and using the distance candidate value, the moving speed candidate value, and the amplitude coefficient candidate value, as the distance estimated value and the moving speed estimated value It is characterized by selection.

  According to the present invention configured as described above, by using the signal coefficient, the size of the detection object and the distance between the human body detection devices are further subdivided, and the distance and the moving speed are estimated. The influence of the size can be reduced to more accurately determine the behavioral state of the person.

  The invention according to claim 3 is a human body detection device using a Doppler signal of a propagation wave reflected by a detection object, and a propagation wave transmitter that radiates a propagation wave toward the detection object; A propagation wave receiving unit for receiving a propagation wave reflected by the detection object, a propagation wave radiated by the propagation wave transmitting unit, and a Doppler signal generated based on the propagation wave received by the propagation wave receiving unit And a term that vibrates a difference between the Doppler signal at time t and the Doppler signal at time t−1 one sampling period Δt before time t at a Doppler frequency that varies with the moving speed of the person, Analysis is performed using a differential analysis physical model that combines a term in which vibration amplitude and a transmission signal change with time, and a plurality of moving speed candidate values of the detection target object and the amplitude of the Doppler signal are increased. A plurality of amplitude intensity candidate values indicating a plurality of amplitude intensity candidate values, estimation means for estimating a plurality of Doppler estimation signals using the moving speed candidate values and amplitude intensity candidate values, the plurality of Doppler estimation signals, and the Doppler A Doppler signal analysis unit that compares the amplitude value with the signal and selects the moving speed candidate value corresponding to one Doppler estimation signal that causes a minimum amplitude value error as the moving speed estimated value and the amplitude intensity candidate value; and And an action state determination unit that determines a person's action state based on the velocity and amplitude intensity of the detection target calculated by the signal analysis unit.

  According to the present invention configured as described above, a calculation is performed by calculating a difference combining a term that vibrates at a Doppler frequency that changes according to a person's moving speed, and a term that changes in vibration amplitude and time. It is possible to simplify, and it is possible to improve the processing speed and determine the accurate behavioral state of a person even with a low-capacity computer

  Further, the invention of claim 4 is the human body detection device according to any one of claims 1 to 3, a urinal body, a solenoid valve for discharging and stopping washing water for washing the urinal body, It is a urinal characterized by having an electromagnetic valve control part which opens and closes an electromagnetic valve based on the action state of a person detected by this human body detection device.

According to the present invention configured as described above, a human behavior state can be accurately grasped with a small amount of computation, and a change in human behavior to be detected can be immediately detected. Can improve the reliability of the cleaning operation.

According to the human body detection device and the urinal provided with the human body detection device of the present invention, it is possible to accurately grasp the human behavior state with a small amount of calculation.
In addition, according to the human body detection device of the present invention and the urinal provided with the same, it is possible to immediately detect a change in the behavior of the person to be detected.

It is a block diagram which shows the whole structure of the urinal by embodiment of this invention. It is a graph which shows an example of the Doppler signal produced | generated by the Doppler signal production | generation part. It is a graph showing an example of the peak frequency and amplitude calculated using the particle filter based on the Doppler signal shown in FIG. It is a graph which shows the value of a Doppler signal, the value of a peak frequency and an amplitude, and the determined person's action, and shows an example of the action which a person approaches. It is a graph which shows the value of a Doppler signal, the value of a peak frequency and an amplitude, and the action of the determined person, and shows an example of the action that a person leaves. It is a flowchart which shows the process by the human body detection apparatus in the urinal of embodiment of this invention. An example of the Doppler signal data in the urinal of the 2nd Embodiment of this invention is shown. It is a graph which shows an example of the result of having estimated the distance between a urinal and a person and the distance between a person and a sensor in the urinal of a 2nd embodiment of the present invention. An example of the Doppler signal data in the urinal of the 3rd Embodiment of this invention is shown. It is a graph which shows an example of the result of having estimated the Doppler signal amplitude intensity and angular frequency in the urinal of the 3rd Embodiment of this invention.

Next, a preferred first embodiment of the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a block diagram showing the overall configuration of a urinal according to an embodiment of the present invention.

  As shown in FIG. 1, a urinal 1 according to an embodiment of the present invention discharges a urinal body 2, a human body detection device 4 built in the urinal body 2, and washing water for washing the urinal body 2. And an electromagnetic valve 6 to be stopped, and an electromagnetic valve control unit 8 for controlling the electromagnetic valve 6.

  In the urinal 1 of the present embodiment, when the human body detecting device 4 detects that a person has stopped approaching the urinal body 2 with the addition of urine, the electromagnetic valve control unit 8 sends a signal to the electromagnetic valve 6, The urinal body 2 is pre-cleaned by opening it for a predetermined time. Further, in the urinal 1, when the human body detecting device 4 detects that the person who has added the urinal has left the urinal body 2, the electromagnetic valve control unit 8 sends a signal to the electromagnetic valve 6, which is opened for a predetermined time. The urinal body 2 is then washed afterwards.

The urinal body 2 includes a bowl portion 2a for receiving urine, a spout 2b provided at the upper portion of the bowl portion for discharging the washing water for washing the bowl portion 2a, and for discharging urine and washing water. And a drain port 2c.
The electromagnetic valve 6 is provided in a water supply pipe connected to the water discharge port 2b, and is configured to be opened and closed by a signal sent from the electromagnetic valve control unit 8.

  When the human body detecting device 4 detects "approach of the human body" indicating that the human body has approached the urinal body 2 with a small amount of use and stopped, the electromagnetic valve control unit 8 receives this detection signal for a predetermined time. The electromagnetic valve 6 is opened and pre-cleaning is performed. Further, when the human body detecting device 4 detects that the human body detecting device 4 has moved away from the urinal body 2 by the human body detecting device 4, the electromagnetic valve 6 receives the detection signal for a predetermined time. Is opened and post-cleaning is performed. Specifically, the electromagnetic valve control unit 8 can be configured by a microprocessor, a memory (not shown), and the like.

  The human body detection device 4 is built in the urinal body 2, and is a microwave transmission unit 10 that is a propagation wave transmission unit that radiates a microwave that is a propagation wave toward a human body that is a detection target, and the microwave transmission unit. The microwave receiving unit 12 which is a propagation wave receiving unit that receives the microwaves radiated from 10 and reflected by the human body, the microwaves radiated by the microwave transmitting unit 10 and the microwaves received by the microwave receiving unit 12 And a Doppler signal generation unit 14 for generating a Doppler signal based on the above.

  Furthermore, the human body detection device 4 analyzes the Doppler signal generated by the Doppler signal generation unit 14 using an autoregressive model, calculates the Doppler signal peak frequency and amplitude, and the Doppler signal. And a behavioral state determination unit 18 that determines the approach of the human body and the withdrawal of the human body, which are human behavioral states, based on the peak frequency and amplitude calculated by the analysis unit 16. In the present embodiment, the human body detection device 4 is built in the urinal body 2, but the human body detection device 4 may be disposed outside the urinal body 2.

  The microwave transmission unit 10 is configured to radiate microwaves from the back side of the urinal body 2 toward the human body approaching the urinal 1. Specifically, the microwave transmission part 10 can be comprised by the microwave oscillator which transmits the microwave of a predetermined frequency at a predetermined time interval. The microwaves radiated from the microwave transmission unit 10 pass through the ceramic urinal body 2 and are radiated toward the front of the urinal body 2.

  The microwave receiving unit 12 is configured to receive the microwave radiated from the microwave transmitting unit 10 and reflected by the human body to be detected. The microwave reflected by the human body passes through the urinal body 2 again and is received by the microwave receiving unit 12 disposed on the back side of the urinal body 2. Specifically, the microwave receiving unit 12 can be configured by a microwave receiver.

  The Doppler signal generation unit 14 is configured to generate a Doppler signal by mixing a part of the microwave radiated from the microwave transmission unit 10 and the microwave received by the microwave reception unit 12 with a mixer. ing. The generated Doppler signal is a signal that contains a lot of frequency components according to the moving speed of the human body reflecting the microwave. The frequency is high when the moving speed is high, and the frequency is low when the moving speed is low. Become.

  The Doppler signal analysis unit 16 analyzes the Doppler signal generated by the Doppler signal generation unit 14 using an autoregressive model, and the peak frequency which is the most abundant frequency component in the analyzed Doppler signal and the peak frequency It is configured to calculate the amplitude of the signal. The peak frequency and amplitude of the Doppler signal are associated with the autoregressive coefficient of the autoregressive model, and the value of the autoregressive coefficient is estimated using a particle filter. Details of the arithmetic processing in the Doppler signal analysis unit 16 will be described later.

  Based on the peak frequency and amplitude calculated by the Doppler signal analysis unit 16, the behavioral state determination unit 18 is configured to determine a human behavioral state such as “approach of human body”, “retreat of human body”, and the like. . The behavior state determination unit 18 determines “approach of the human body” and “retreat of the human body” based on the fluctuation tendency of the peak frequency and the amplitude within the predetermined period. Details of the calculation processing in the behavior state determination unit 18 will be described later. Specifically, the Doppler signal generation unit 14, the Doppler signal analysis unit 16, and the behavior state determination unit 18 may be configured by an A / D converter, a memory, a microprocessor, a program for operating the A / D converter, and the like. it can.

Next, processing executed by the Doppler signal analysis unit and the behavior state determination unit according to the embodiment of the present invention will be described with reference to FIGS.
FIG. 2 is a graph illustrating an example of the Doppler signal generated by the Doppler signal generation unit. FIG. 3 is a graph showing an example of the peak frequency and amplitude calculated using the particle filter based on the Doppler signal shown in FIG. FIG. 4 is a graph illustrating an example of sampled Doppler signal values, peak frequency and amplitude values calculated based on the values, and human behavior determined by the behavior state determination unit. FIG. 5 is a graph showing the value of the Doppler signal, the peak frequency and the amplitude, and the determined human behavior, and shows an example of the behavior of the person leaving.

  First, when a person approaches the urinal 1 from a distance and stops in the vicinity of the urinal 1, the Doppler signal generator 14 outputs a Doppler signal as shown in FIG. Note that the horizontal axis in FIG. 2 represents time series, the numerical value on the horizontal axis represents the number of steps for sampling data, and the vertical axis represents the amplitude of the Doppler signal. In FIG. 2, the Doppler signal is sampled at an interval of 1/1024 sec, and the graph of FIG. 2 shows the change of the Doppler signal for a period of about 3 sec. 2 shows data sampled at an interval of 1/1024 sec. However, the number of data actually required for analysis in the Doppler signal analysis unit 16 is significantly smaller than this, and at a coarser interval than in FIG. The Doppler signal can be sampled.

The Doppler signal analysis unit 16 is configured to model and analyze the Doppler signal shown in FIG. 2 by a second-order non-stationary autoregressive model shown in Equation 1.

Is an estimated value of the Doppler signal at the current time, and a _i (t) is an autoregressive coefficient that changes with time. Further, x (k−i) is a detected Doppler signal value at a time before i steps, and ε (k) is an observation error.

Furthermore, the relationship of the following formulas 2 and 3 is established between the two autoregressive coefficients a ₁ (t) and a ₂ (t), the peak frequency ω of the signal, and the amplitude r at the peak frequency. It is known to do.

  The Doppler signal analysis unit 16 uses the particle filter to detect the signal peak frequency ω and the amplitude r at the peak frequency, and detects Doppler signal values x (k) and x (k−1) for three steps. , X (k−2).

  Further, in the present embodiment, the Doppler signal analysis unit 16 calculates the peak frequency ω and the amplitude r based on the three Doppler signal values x (k), x (k−1), and x (k−2). A particle filter is used as a technique for this. The particle filter generates a plurality of combinations of peak frequencies and amplitudes as particles within a range assumed in advance, and sets the particles and Doppler signal values x (k−1) and x (k−2). Use to calculate the Doppler signal estimate at the current time. Next, the plurality of calculated Doppler signal estimated values are compared with the actually measured Doppler signal value x (k) at the current time, and the actual Doppler signal value x (k) is compared with the closest Doppler signal estimated value. The given particle is determined as an estimate of the peak frequency ω and amplitude r.

  Since the amount of calculation required for calculating the peak frequency ω and the amplitude r using the particle filter is relatively small, in this embodiment, after measuring the Doppler signal value at the current time, the next Doppler signal value is measured. In the meantime, the peak frequency ω and the amplitude r can be calculated.

  FIG. 3 shows a graph of peak frequency ω and amplitude r calculated in time series by the particle filter using the autoregressive model in the Doppler signal analysis unit 16 based on the Doppler signal shown in FIG. ing. In the example illustrated in FIG. 3, the values of the peak frequency ω and the amplitude r are estimated in a time series at intervals of about 16 msec with respect to Doppler signal values sampled at intervals of about 16 msec. In the human body detection device 4 in the present embodiment, it is possible to obtain the estimated values of the peak frequency ω and the amplitude r at a sufficiently short time interval without extremely shortening the sampling interval of the Doppler signal as described above ( In the invention described in Patent Document 1, one peak frequency ω and amplitude r are calculated for 128 Doppler signals.) Thereby, it becomes possible to grasp the fluctuation tendency with respect to time of the peak frequency ω and the amplitude r at a practical time interval.

  From the graph of FIG. 3, the peak frequency ω increases gradually when a distant person approaches the urinal 1 (time around 0 to 1000), and the peak frequency decreases when the speed of walking closer to the urinal 1 is reduced. ω decreases (around time 2000 to 2500), and when it stops in front of the urinal 1, the tendency that the peak frequency ω becomes a low value (around time 3000) can be read. In addition, the amplitude r tends to increase as the person reaches the vicinity of the urinal 1 (time around 2000 to 3000).

  Furthermore, the graph shown in FIG. 5 shows an example of a Doppler signal when a person who is stationary near the urinal 1 moves out of the urinal 1 vicinity. As shown in FIG. 5, the temporal transition of the Doppler signal and its peak frequency and amplitude is opposite to that in FIG.

  The behavior state determination unit 18 compares the peak frequency ω and the amplitude r calculated by the Doppler signal analysis unit 16 with a predetermined standard stored in a memory (not shown) so as to determine the human behavior state. It is configured.

  As shown in FIG. 4, in this embodiment, the behavior state determination unit 18 has a peak frequency ω of 1.4 [rad] or more from 5 sampling steps before the current time to the current time from 5 sampling steps before. When the amplitude r is 0.65 or more, and the fluctuation tendency of the peak frequency ω between 5 sampling steps and the current time is negative, the person is close to the urinal 1 Judge that approached. In the example illustrated in FIG. 4, the behavior state determination unit 18 determines that the person's behavior state is “approach” to the vicinity of the urinal at time 2176.

  The fluctuation tendency means a value obtained by dividing the amount of change in peak frequency or amplitude by the period in which the change occurs. For example, the fluctuation tendency of the peak frequency ω at the current time can be calculated by subtracting the peak frequency one sampling step before the peak frequency at the current time and dividing the value by the sampling interval.

  In addition, the behavioral state determination unit 18 has a peak frequency ω of 0.7 [rad] or less at the current time and an amplitude r of 0.75 or more from 5 sampling steps to the current time, and from 5 sampling steps before. When the fluctuation tendency of the peak frequency ω until the current time is negative and the absolute value of the fluctuation tendency of the amplitude r between the five sampling steps and the current time is 0.0025 or less, the person is small. It is determined that it is stationary in the vicinity of the toilet 1, that is, it is added to a small use. In the example illustrated in FIG. 4, the behavior state determination unit 18 determines that the human behavior state is “still” at time 2720.

  Furthermore, as shown in FIG. 5, the behavioral state determination unit 18 has a peak frequency ω of 1.4 [rad] or more between 5 sampling steps and the current time, and is between 5 sampling steps and the current time. When the amplitude r is 0.65 or more and the fluctuation tendency of the peak frequency ω from 5 sampling steps to the current time is positive, it is determined that the person has left the vicinity of the urinal 1. In the example illustrated in FIG. 5, the behavior state determination unit 18 determines that the person's behavior state is “withdrawal” at time 428.

  Next, the operation of the urinal 1 according to the embodiment of the present invention will be described with reference to FIG. FIG. 6 is a flowchart showing processing by the human body detection device 4 in the urinal of the present embodiment.

  First, in step S1 of FIG. 6, when the human body detection device 4 is powered on, the process proceeds to step S2 and enters a standby mode. In this standby mode, the microwave transmission unit 10 radiates microwaves in front of the urinal 1, and the microwave reception unit 12 receives microwaves. The Doppler signal generation unit 14 generates a Doppler signal based on the radiated microwave and the received microwave. Further, the Doppler signal analysis unit 16 analyzes the generated and sampled Doppler signal and sequentially calculates the peak frequency ω and the amplitude r.

  Next, in step S3, it is determined whether or not a person has approached the vicinity of the urinal 1. That is, the behavior state determination unit 18 determines whether or not a person has approached based on the calculated peak frequency ω and amplitude r. The behavior state determination unit 18 determines whether or not the following three conditions (a) to (c) are satisfied. (A) The peak frequency ω between 5 sampling steps and the current time is 1.4 [rad] or more. (B) The amplitude r between 5 sampling steps and the current time is 0.65 or more. (C) The fluctuation tendency of the peak frequency ω between 5 sampling steps and the current time is negative.

  When these three conditions (a) to (c) are satisfied, the behavior state determination unit 18 determines that the person's behavior state is “approach” to the vicinity of the urinal, and the processing is step. Proceed to S4. If the three conditions are not satisfied, the process returns to step S2 and the standby state is maintained.

  In step S4, it is recognized that the action state of the person is “approaching”. Further, in step S5, it is determined whether or not the person approaching the urinal 1 is stationary and starts to use. That is, the behavior state determination unit 18 determines whether or not the following four conditions (d) to (g) are satisfied. (D) The peak frequency ω at the current time is 0.7 [rad] or less. (E) The amplitude r between 5 sampling steps and the current time is 0.75 or more. (F) The fluctuation tendency of the peak frequency ω between 5 sampling steps and the current time is negative. (G) The absolute value of the fluctuation tendency of the amplitude r between 5 sampling steps and the current time is 0.0025 or less.

  When these four conditions (d) to (g) are satisfied, the behavioral state determination unit 18 determines that the person approaching the urinal vicinity has started to use, and the process proceeds to step S6. move on. On the other hand, after the approach state is recognized in step S3, if it is not determined that the person has started to use even after a predetermined time has elapsed, the process returns to step S2 and enters a standby state.

  In step S6, the solenoid valve control unit 8 sends a control signal to the solenoid valve 6 to open the solenoid valve 6 for a predetermined time. Thereby, washing water is discharged from the water outlet 2b of the urinal body 2, and the bowl part 2a is pre-washed.

  Next, in step S <b> 7, it is determined whether or not the person who has started occupancy has left the urinal 1. That is, the behavior state determination unit 18 determines whether the following four conditions (h) to (j) are satisfied. (H) The peak frequency ω between 5 sampling steps and the current time is 1.4 [rad] or more. (I) The amplitude r between the 5 sampling steps and the current time is 0.65 or more. (J) The fluctuation tendency of the peak frequency ω between 5 sampling steps and the current time is positive.

  If these three conditions (h) to (j) are satisfied, the behavior state determination unit 18 determines that the person who started the small use has left, and the process proceeds to step S8. On the other hand, when the three conditions are not satisfied, the determination in step S7 is repeated.

  In step S8, the solenoid valve control unit 8 sends a control signal to the solenoid valve 6 to open the solenoid valve 6 for a predetermined time. Thereby, the wash water is discharged from the water outlet 2b of the urinal body 2, and the bowl portion 2a is post-washed.

  Next, the urinal 1 in the second embodiment will be specifically described with reference to the drawings. In the urinal 1 in the first embodiment, Doppler signal processing is performed using the relationship between the autoregressive model of Formula 1 and Formulas 2 and 3, but in the urinal of the second embodiment, the Doppler signal analysis unit Reference numeral 16 is configured to model and analyze the Doppler signal shown in FIG. 7 using the Doppler analysis physical model shown in Equations 4 and 5.

Where t is the time, v (t) is the movement speed of the person at time t, x ₀ is the distance at which the sensor has started detection (the speed in the direction in which the person approaches is positive), and c is the microwave The velocity of propagation in air, f is the frequency of the microwave sensor, A is a constant, and K is a signal coefficient.

Using Equation 4 and Equation 5, using the same particle filter technology as the first embodiment, when such is estimated constants A, and K and human velocity V (t) the distance x _0, human movement speed and sensor Can be estimated.

  FIG. 8 shows an estimation result of the movement speed and distance of the person who estimated the Doppler signal of FIG. 7 in the second embodiment. As shown in FIG. 8, if the movement speed of the person and the distance between the sensor and the person (the sensor is fixedly installed in the urinal 1 and corresponds to the distance between the urinal and the person) can be estimated, the movement speed and the urinal and the person can be estimated. A threshold is set in advance for each of the distances. If the distance between the urinal and the person is smaller than the distance threshold, the person approaches the urinal, and if the moving speed is smaller than the speed threshold, the person stops in front of the urinal. It is determined that the user is trying to detect the presence of the user. Further, when the distance between the person and the urinal is larger than the distance threshold and the movement speed of the person is larger than the speed threshold, it is determined that the person is separated from the urinal, and is the same as the urinal of the first embodiment. Control the urinal solenoid valve. In the second embodiment, since the distance between the moving body and the urinal is estimated, it can be determined that the person is approaching the urinal more reliably. In addition, since the constant K is estimated by the particle filter, it is possible to estimate according to the change in the signal intensity affected by the size of the human body, etc., and to reduce the influence of the size of the detection object. An accurate behavior state of a person can be determined.

  Next, the urinal 1 according to the third embodiment will be specifically described with reference to the drawings. In the urinal 1 in the first embodiment, the Doppler signal processing is performed using the relationship between the autoregressive model of Formula 1 and Formula 2 and Formula 3, but in the urinal of the third embodiment, the Doppler signal analysis unit Reference numeral 16 is configured to analyze the Doppler signal shown in FIG. 9 by modeling it using a differential analysis physical model shown in Equation 6.

Here, t is the time, ω (t) is the Doppler angular frequency corresponding to the moving speed of the person at time t, Δω is the amount of change in the number of angular cycles in the difference section, and c (t) is the signal amplitude. If c (t) and ω (t) are estimated using Equation 6 and the same particle filter technique as in the first embodiment, the human moving speed v (t) can be estimated by Equation 7.

FIG. 10 shows the estimation results of the movement speed and signal amplitude of the person who estimated the Doppler signal of FIG. 9 in the third embodiment.
Since c (t) is the Doppler signal amplitude intensity, it increases when a person approaches the sensor and decreases when the person moves away from the sensor. Therefore, a threshold value is set for each of c (t) and ω (t), and if c (t) is equal to or greater than the amplitude intensity threshold value and ω (t) is equal to or less than the angular velocity threshold value, the person approaches the urinal. If it is determined that the vehicle is about to stop in front of the urinal, and ω (t) and c (t) are both equal to or greater than the threshold and less than the threshold, it is determined that the person is away from the urinal. Control the urinal solenoid valve similar to the urinal of the form. In the third embodiment, among the differential analysis physical models expressed by Equation 6, there are two estimation targets, c (t) and ω (t), which can be estimated with a smaller amount of calculation than in the second embodiment. Can be determined.

  As mentioned above, although preferable embodiment of this invention was described, a various change can be added to embodiment mentioned above. In particular, in the above-described embodiment, the human body detection device of the present invention is applied to a urinal, but the human body detection device of the present invention can be applied to various devices such as an automatic faucet and a flush toilet. it can.

  In the first embodiment described above, the Doppler signal analysis unit calculates the peak frequency and amplitude using a second-order autoregressive model. However, a third-order or higher-order autoregressive model is applied to the present invention. can do. For example, the peak frequency and amplitude can be calculated using a Doppler signal for four samples in a third-order autoregressive model and using a Doppler signal for five samples in a fourth-order autoregressive model.

  Furthermore, in the first embodiment described above, the behavior state determination unit refers to the peak frequency and amplitude from the current time to 5 sampling steps before, but appropriately changes the period for referring to the peak frequency and amplitude. be able to.

  In the above-described embodiment, the crisp expression, which is a clear numerical value, is used as a reference for the peak frequency, amplitude, distance, moving speed, and fluctuation tendency of the behavior state determination unit that determines the human behavior state. The fuzzy number can also be applied to this.

In the third embodiment described above, the Doppler signal analysis unit uses the difference between the Doppler signal Y _{t at} time _t and the Doppler signal Y _{t + Δt} after the time Δt that is the sampling period. The peak frequency and the amplitude can be calculated using the difference of the Doppler signal having a time difference that is an integral multiple of the sampling period instead of the difference for each sampling period.

In the second embodiment described above, the movement speed of a person and the distance between the person and the urinal, and in the third embodiment, the movement speed of the person and the signal amplitude intensity are respectively set at predetermined thresholds. However, it is also possible to determine a person's behavior based on a fluctuation tendency within a predetermined period.

  Furthermore, in the above-described embodiment, microwaves are used as propagating waves radiated toward the detection target. However, measurement using the Doppler effect such as electromagnetic waves other than microwaves, laser light, and ultrasonic waves is possible. Any propagating wave can be used.

DESCRIPTION OF SYMBOLS 1 Urinal by embodiment of this invention 2 Urinal body 2a Bowl part 2b Water outlet 2c Drainage port 4 Human body detection apparatus 6 Electromagnetic valve 8 Electromagnetic valve control part 10 Microwave transmission part (propagation wave transmission part)
12 Microwave receiver (propagating wave receiver)
14 Doppler signal generation unit 16 Doppler signal analysis unit 18 Action state determination unit

Claims (4)

  Propagation wave transmitter for radiating a propagation wave toward the object to be detected; Propagation wave receiver for receiving a wave reflected by the object to be detected; Propagation wave radiated by the propagation wave transmitter and the propagation In a human body detection device comprising: a Doppler signal generation unit that generates a Doppler signal based on a propagation wave received by a wave reception unit; a term that vibrates at a Doppler frequency that varies depending on a moving speed of the detection object And a particle filter that analyzes by a Doppler analysis physical model that combines a DC component of the sensor signal, and a term in which the phase difference between the transmission signal and the reflection signal varies with time, and a plurality of particles between the detection object and the human body detection device A plurality of candidate distance values and a plurality of candidate moving speed values of the detection object, and using the candidate distance values and the candidate moving speed values, The estimation unit for estimating a Doppler estimation signal, the plurality of Doppler estimation signals, and the amplitude value of the Doppler signal are compared, and the distance candidate corresponding to one Doppler estimation signal whose difference is the smallest amplitude value error A Doppler signal analysis unit that selects a value and the moving speed candidate value as a distance estimated value and a moving speed estimated value, and a human action state based on the distance and moving speed of the detection target calculated by the Doppler signal analyzing unit. A human body detection device comprising: an action state determination unit for determining.   The human body detection apparatus according to claim 1, wherein the Doppler analysis physical model includes a signal coefficient that multiplies a term of Doppler effect and a term of beat, and the Doppler signal analysis unit includes a plurality of degrees of amplification of the Doppler signal. Human body detection, wherein the distance candidate value, the moving speed candidate value, and the amplitude coefficient candidate value are selected as the distance estimated value and the moving speed estimated value. apparatus. A human body detection device using a Doppler signal of a propagation wave reflected by a detection object, a propagation wave transmitter for radiating the propagation wave toward the detection object, and a propagation wave reflected by the detection object , A Doppler signal generation unit that generates a Doppler signal based on the propagation wave radiated by the propagation wave transmission unit and the propagation wave received by the propagation wave reception unit, and a Doppler at time t The difference between the signal and the Doppler signal at time t−1 one sampling period Δt before time t, the term that vibrates at the Doppler frequency that changes at the moving speed of the person, the vibration amplitude, and the transmission signal change over time. Analyzing using a differential analysis physical model combining terms, a plurality of candidate moving velocity values of the detection object, and a plurality of amplitude intensity candidates indicating the magnitude of the amplitude of the Doppler signal And estimating means for estimating a plurality of Doppler estimation signals using the moving speed candidate value and the amplitude strength candidate value, comparing the amplitude values of the plurality of Doppler estimation signals and the Doppler signal, A Doppler signal analysis unit that selects the moving speed candidate value corresponding to one Doppler estimation signal that causes a minimum amplitude value error as a moving speed estimation value and an amplitude intensity candidate value, and a detection target calculated by the Doppler signal analysis unit A human body detection apparatus comprising: an action state determination unit that determines a person's action state based on the speed and amplitude intensity of an object.   A urinal body, an electromagnetic valve for discharging and stopping washing water for washing the urinal body, the human body detection device according to any one of claims 1 to 3, and a person detected by the human body detection device And a solenoid valve control unit that opens and closes the solenoid valve based on the action state of the urinal.